Dynamic Redeployment of Nurses Across Hospitals: A Sample Robust Optimization Approach

Wei Liu, Tianchun Li, Mengshi Lu, Pengyi Shi · 2025

Read Dynamic Redeployment of Nurses Across Hospitals: A Sample Robust Optimization Approach by Wei Liu, Tianchun Li, Mengshi Lu, Pengyi Shi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-07387.txt.

Introduction

Wei Liu and 3 co-authors study how hospital networks can move nurses from sites with spare staff to sites facing shortages in a 2025 paper. The practice, called internal travel, grew during the COVID-19 pandemic. It needs advance notice and short secondments, and demand forecasts from old data prove unreliable when conditions change fast, which limits plain sample-based planning.

The authors frame the problem as a stochastic dynamic program, a model that decides step by step under uncertainty, guarded by a sample robust optimization approach. A linear decision rule turns the model into a tractable linear program a computer can solve. Connectivity between hospitals has the largest effect, cutting cost, redeployments, and travel distance. Its benefits, though, depend on matching how long secondments last to the shape of the network.

Catalog record

CREATORWei Liu, Tianchun Li, Mengshi Lu, Pengyi Shi
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE120 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-07387

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Dynamic Redeployment of Nurses Across Hospitals: A Sample Robust Optimization Approach Wei Liu, Tianchun Li, Mengshi Lu, Pengyi Shi Abstract Problem definition: We study a workforce redeployment problem in hospital networks, where clinical staff, such as nurses, are temporarily reassigned from overstaffed to understaffed sites to address short-term imbalances. This practice of “internal travel,” which gained traction during the COVID-19 pandemic to tackle nurse shortages, presents new operational challenges that require tailored analytical support. Key requirements such as advance notice and short-term secondments must be incorporated. Moreover, in rapidly evolving environments, reliance on historical data leads to unreliable forecasts, limiting the effectiveness of traditional sample-based methods. Methodology: We formulate the problem as a stochastic dynamic program and incorporate demand uncertainty via a sample robust optimization (SRO) framework. Using linear decision rule approximation, we reformulate the problem as a tractable linear program. Results: We evaluate the impact of key network design components on system performance. Network connectivity has the largest effect in reducing the total cost, number of redeployments, and travel distance, but its benefits depend on aligning the secondment duration with the network structure. Full connectivity without proper secondments can be counterproductive. The SRO approach outperforms the traditional sample-average method in the presence of demand surges or under-forecasts by better anticipating emergency redeployments. Managerial implications: Internal travel programs offer a promising strategy to alleviate workforce shortages in healthcare systems. Our results highlight the importance of network design, aligning secondment durations with the network structure, and adopting planning methods that are robust to demand surges or inaccurate predictions. \AppendGraphicsExtensions .pdf keywords workforce management, robust nurse staffing, inter-hospital transfer † † manuscriptno: MSOM-2025 † † authors: Faculty of Business in Scitech, School of Management, University of Science and Technology of China, [email protected] Elmore Family School of Electrical and Computer Engineering, Purdue University, [email protected] Mitchell E. Daniels, Jr. School of Business, Purdue University, [email protected] Mitchell E. Daniels, Jr. School of Business, Purdue University, [email protected] † † affiliation: † † affiliation: † † affiliation: † † affiliation: † † affiliation: † † affiliation: † † affiliation: † † affiliation: 1 Introduction The decades-long healthcare workforce shortage has escalated into a global crisis, with the United States projected to face significant deficits across multiple frontline roles—including nurses, nurse assistants, and clinical technicians—by 2030 ( Lyons 2023 , Hoover et al. 2024 and Murphy 2024 ). These shortages have serious implications, including increased staff burnout, higher risks of medical errors, and adverse patient outcomes ( University of St. Augustine for Health Sciences 2024 , Carayon and Gurses 2018 ). In response, a growing number of hospital systems have adopted “internal travel” programs, which temporarily redeploy medical staff across hospital sites to manage demand imbalances ( Brenner 2025 , and Yuan 2025 ). This practice, which emerged during the COVID-19 pandemic, introduces new operational challenges that demand tailored analytical support. Our community partner, Indiana University Health (IU Health), is one such system that has implemented the workforce redeployment initiative, termed the Delta Coverage program ( Helm et al. 2024 ). Motivated by this context, our research in this paper focuses on nurse redeployment as a representative case, with the problem setup applicable to other roles such as nurse assistants and technicians. Next, we introduce the background and outline the key analytical challenges using IU Health as a motivating example, while highlighting the relevance in the broader healthcare context. We then present an overview of our approach and key contributions. Figure 1 : Locations of each hospital in the IU Health network. 1.1 Problem Background and Operations Challenges IU Health is the largest healthcare system in Indiana, operating 16 hospitals. Figure 1 shows the distribution of these hospitals throughout the state, with four main regions segregated based on their proximity to Indianapolis ( IU Health 2023 ). The central regions, including Indianapolis and its suburban areas, have higher population densities. During the early stages of the COVID-19 pandemic, they experienced more confirmed cases and hospitalizations (per 10,000 residents, Indiana State Government 2023 ), and hence, faced severe nurse shortages at the onset of the outbreak. In contrast, rural regions with lower population densities initially saw fewer cases but faced their unique challenges. With smaller staffing pools and limited surge capacity, these hospitals struggled to absorb day-to-day demand fluctuations. When COVID-19 eventually spread to these areas, they experienced even more severe staffing shortfalls than their urban counterparts. This stochastic and regionally asynchronous demand pattern was a key driver behind IU Health’s implementation of the nurse redeployment program. While the idea behind workforce redeployment is intuitive (i.e., resource pooling), its implementation raises several key operations management (OM) challenges, including network design, advance-notice deployment decisions, and planning under demand uncertainty. We elaborate on each of these elements below, which is broadly relevant to hospital systems facing similar staffing shortages. The first element is network design. Given the statewide distribution of hospitals, two structures are possible. The current model at IU Health follows a hub-and-spoke design ( Števárová and Badánik 2018 ), where nurses travel between central and rural hospitals and return home the same day. An alternative is a fully connected (or point-to-point) network ( Cook and Goodwin 2008 ), allowing deployments between any hospitals, including rural-to-rural ones. IU Health has been actively exploring this design to increase flexibility and better respond to asynchronous demand surges across regions. However, longer distances between rural sites make daily commuting impractical, which leads to the second design element. The second element involves accounting for long-distance redeployments through advance notice and secondments. Secondment refers to a multi-day assignment at the receiving hospital, during which the nurse remains on-site to avoid prolonged and potentially hazardous daily commutes. It often comes with additional pay to compensate for remote travel. IU Health initially implemented a real-time redeployment model, where decisions were made on the day of need, e.g., at 5 am each day before the morning shifts start. However, participation was minimal, as nurses were unwilling to engage without adequate notice. This led to the development of a two-stage framework: during the on-call planning phase (two weeks in advance), tentative assignments are made to provide nurses with early notification; during the deployment phase (each morning), these plans can be revised, either by canceling the planned deployment (with a cancellation fee) or initiating emergency deployment (at a premium cost). This redesign highlights the human-centric nature of the problem: nurses are not items, unlike in traditional inventory transshipment or physical resource repositioning models (e.g., Shu et al. 2013 , and He et al. 2020 ). Operational plans must incorporate advance notice, consider the travel burden, and account for the cost of modifying the plans. Nurses are also not interchangeable physical resources—–each nurse has a home location. Nurses need to return to their home location after the secondment, making this fundamentally different from a network flow problem. Detailed discussion on the difference from the transshipment and repositioning literature is in Section 2 . Secondment further complicates the problem structure by introducing temporal dependencies within the deployment phase; that is, a decision to deploy a nurse today affects their availability for multiple future days. This necessitates a multi-stage decision problem with interdependent constraints across time. The resulting complexity exceeds what ad hoc

Read the complete text